In 2025, the heat around artificial intelligence-driven glasses is unmistakable. Hard to believe, considering that just a year ago, the smart glasses market had been all but abandoned, kept alive mostly by a handful of state-backed investors.
“By 2025, strategic investors and market-driven funds are all here,” said Yang Longsheng, CEO of Inmo, a company widely recognized as one of China’s rising augmented reality (AR) players. He recalled that its latest financing round was closed in just two months.
After meeting with more than a hundred investors, Inmo secured RMB 150 million (USD 21 million) in Series B2 funding. The round was co-led by Puhua Capital, Liangxi Industry Development Group, and 58.com-backed Pegasus Capital.
Yang, who once led Coolpad’s smart wearables division before striking out on his own, has seen the ups and downs of the industry firsthand. Now, with the rise of foundation models from OpenAI and DeepSeek fueling new interest in AI applications, the sector is seeing a surge in demand, and AI glasses are emerging as one of the most promising interfaces.
Glasses are uniquely suited for AI integration. They are worn close to the eyes, naturally positioned for information display, and already a familiar, widely adopted form factor. It’s no surprise they have become a hotbed for innovation.
The latest boom began in 2024 with the runaway success of Meta’s AI-powered glasses made in collaboration with Ray-Ban. The product sold more than two million units that year. Following that success, tech giants and startups quickly stepped up to claim a stake.
Inmo beat Xiaomi to the punch by launching its Air3 glasses in November 2024. But unlike many of its peers, Inmo isn’t shifting away from AR. Instead, AR remains its core focus. The Air3 supports 1080p resolution, binocular full-color display, and includes AI-powered semantic interaction, spatial computing, and multi-screen collaboration.
Xiaomi’s AI glasses, released in late June, sent the market into overdrive. According to Yang, traffic across the entire smart glasses market has surged tenfold since that launch. Industry-wide sales this year are expected to be two to three times higher than last year.
Three types of contenders
The AI glasses market in China is now populated by three distinct types of players.
First are the major internet companies like Xiaomi and Baidu. Second are the AR-native firms which include Inmo, Xreal, Rokid, and RayNeo, who are now expanding into the space. The third group comprises new entrants with no prior experience in eyewear: consumer electronics upstarts like Sharge and a bevy of white-label manufacturers from Shenzhen’s Huaqiangbei district.
In China, building a glasses-shaped camera is relatively easy, which is why so many players with radically different backgrounds are jumping in. Compared to other hardware, smart glasses come with significantly lower technical barriers. But Yang warned that this may be a market rife with pitfalls.
Where the real challenges lie
According to Yang, future competition in the AI glasses space will play out on three fronts.
- Hardware: Can smart glasses achieve true miniaturization and comfort, making them wearable all day like regular glasses? This is key to long-term adoption. Yang also pointed to optics and display technology as critical elements. Accordingly, manufacturers must deliver better yields, lower costs, and higher visual quality.
- AI capabilities: If every player just plugs into the same generic foundation models, products will quickly become indistinguishable. The real challenge, Yang said, is using AI as a middleware layer, connecting hardware to real-world scenarios, enabling better data capture, and unlocking spatial computing.
- Supply chain: Reducing production costs will determine how quickly AI glasses can go mainstream.
At this early stage, few companies have reached product maturity, achieved meaningful user validation, or found a clear commercial path. “Big tech companies are still testing the waters. Startups probably have two or three years of runway to build something enduring,” Yang said.
The following transcript has been edited and consolidated for brevity and clarity.
Xiaomi enters the fold
36Kr: What is Xiaomi’s impact on the AI glasses market?
Yang Longsheng (YL): Xiaomi’s entry is great for the industry. AI glasses used to be niche, but now the category is getting real attention. It’s improving public awareness and educating the market.
Looking at traffic data, during its launch window, Xiaomi’s entry lifted overall industry traffic by nearly ten times. On a daily basis, I’d estimate it’s brought about a threefold increase. For many companies, annual sales could double or even triple this year. For us, sales have increased by about 50% since Xiaomi’s release.
36Kr: You don’t view Xiaomi as a threat?
YL: Not really. Right now, there’s a clear distinction between what startups and bigwigs like Xiaomi are offering. We’re not yet in a stage where market share is being cannibalized.
36Kr: What are the cost components of a pair of smart glasses? Let’s use Xiaomi as an example.
YL: Xiaomi’s model represents the most basic form factor for AI glasses. The biggest cost component is the chip. Qualcomm’s AR1 alone costs about USD 44. With the rest of the mainboard, it comes to around RMB 600–700 (USD 84–98). Add the optical and sensor components, and you’re looking at RMB 800–900 (USD 112–126). Including the casing and structural parts, the total bill of materials (BOM) hits roughly RMB 1,200 (USD 168).
Our glasses are a step up through the inclusion of a display module. But we’ve kept BOM costs comparable to camera-only models by localizing most components. Our display-enabled, AI-powered glasses have BOM costs around RMB 1,700–1,800 (USD 238–252).
RELATED ARTICLE
Market frenzy
36Kr: Was this your largest funding round to date?
YL: Yes. This Series B2 round was worth RMB 150 million. Our broader Series B round was divided into three parts and should total close to RMB 300 million (USD 42 million).
36Kr: The smart glasses space in China has seen many ups and downs. What’s fueling this new boom?
YL: OpenAI and DeepSeek reignited AI fever. That wave of enthusiasm spilled into hardware. AI needs tangible interfaces to connect with users, and glasses are among the most intimate, immediate devices available.
36Kr: A few years ago, only state-backed investors were interested. How different is it today?
YL: Completely different. Last year, state-owned capital was basically the only game in town. But this year, investors and market funds started pouring in. Thanks to referrals from existing backers and friends, we were approached by over 100 firms. We started serious discussions in April or May, and the whole transaction closed in just over two months.
36Kr: Any specific moment that made you realize how hot the market’s gotten?
YL: For one, friends who never used to care about smart glasses are suddenly asking if Xiaomi’s glasses are worth buying. Also, several top internet companies might roll out products by the end of this year.
36Kr: What are investors most curious about?
YL: They always ask three questions:
- What’s your view of the AI glasses market?
- What impact will Xiaomi have?
- And how will startups compete with the tech giants?
A window for startups
36Kr: Inmo is sticking with AR features, unlike others that are downplaying them. Why?
YL: We believe AI glasses will inevitably evolve to include displays. If you look at the market, audio-based smart glasses without displays hold only a tiny market share. Consumers expect cameras and displays. That’s their baseline vision of what AI glasses should be.
36Kr: How should startups position themselves now that big players are involved?
YL: The tech giants are just dipping their toes in. That gives startups two to three years to become dominant brands. In terms of market share and product sophistication, we’re still ahead. The market is already polarizing, and investors are backing the top two or three players.
36Kr: Who’s buying your glasses?
YL: We focus on new middle-class consumers in first- and second-tier cities. These are generally style-conscious buyers who favor differentiated, premium experiences. They are different from Xiaomi or Huawei’s base.
36Kr: What is Inmo’s moat?
YL: Several key areas. First, optics: we were the first in China to mass-produce waveguide-based AR glasses with full-color binocular displays and a large field of view.
Second, the operating system. We’ve integrated AI deeply into the middleware layer, which supports native use cases like ordering food or shopping directly via voice commands. We’re also exploring social and entertainment applications.
Third, supply chain. Our localization rate is now above 80%, giving us a strong cost advantage. Our AI glasses are now priced close to camera-only models. We also partnered with Asia Optical to open our first branded offline store.
36Kr: Has hardware cost come down overall?
YL: Yes, by about 30% compared to earlier iterations. We’ve improved yield rates by 15% through better design and material processing for array waveguides.
36Kr: How big is the market now, and where do you stand?
YL: 2025 is shaping up to be much stronger than past years. We estimate China will ship 600,000–700,000 units, nearly three times last year’s figure.
Globally, several million units will be sold, with Meta capturing the lion’s share. Xiaomi is expected to ship 300,000–400,000 units. We’re on track to ship about 150,000 units, likely the second highest in China.
Minimal barriers, many traps
36Kr: AI glasses seem easy to replicate, and Huaqiangbei vendors are jumping in. Is it really that easy?
YL: Building a camera that looks like glasses? Sure, that’s simple. But delivering intuitive AI functionality and smooth interaction. That’s the hard part.
Take Xiaomi’s product. It took them more than a year to develop. Every function involves deep optimization. For example, to support real-time AI translation, we had to address latency and create algorithms that accurately capture and translate what you’re saying. Even a basic phone call requires solving directional microphone issues.
Then there’s battery life. We use a dual-chip architecture comprising low-power Bluetooth for everyday tasks, and high-performance system-on-chip for processing images and speech. That introduces challenges with communication protocols. It’s all about the details, the real R&D-heavy work. That’s why white-label vendors can’t keep up. Most of their so-called AI glasses are just surveillance cameras disguised as eyewear.
36Kr: Right now, many AI glasses feel similar. Why is that?
YL: It’s early days. Most players are solving the same few problems: translation, voice assistants, basic tasks. As technology evolves, we’ll see more diverse use cases. Eventually, everything a smartphone app does could be reimagined for glasses.
For developers like us, AI is about rethinking the OS and the interaction model. We’re embedding AI in the middleware to reduce friction and speed up access to information. That’s the real value AI brings. But we don’t intend to build foundation models ourselves. That’s not our role.
36Kr: Where are the real long-term barriers to entry?
YL: There are three key areas.
- First, miniaturization. Can we make glasses that are always on, as light and wearable as prescription eyewear?
- Second, display quality. We’ve invested heavily in waveguide optics to raise performance standards. Third, system-level AI, focusing on enhancing spatial computing and interaction through middleware and six degrees-of-freedom capabilities.
- And finally, the supply chain. We’re pushing to raise our localization rate to 90% or even 100%, so prices can drop to smartwatch or earbud levels. That’s how AI glasses will break into the mass market.
36Kr: What’s next for Inmo in the second half of the year?
YL: We’re investing in all three of those priorities. First, we’ve built a new waveguide production line to improve yield and quality. That alone took RMB 40–50 million (USD 5.6–7 million).
Second, on the AI side, we’re redesigning our OS middleware to shorten user interaction flows. For instance, if you want to order takeout, the glasses can suggest options directly without requiring any searching.
We’ve also launched a RMB 50 million (USD 7 million) fund to incubate native agentic AI applications that will enrich the smart glasses ecosystem.
36Kr: People have long asked what the “killer app” for smart glasses might be. Has AI changed your answer?
YL: AI isn’t an application. It’s a foundational layer. But it will lead to those apps you’re alluding to. Right now, tool-based features like translation and ordering food offer real value. Compared to taking out a phone, glasses get the job done faster.
In the longer term, I believe the big breakthroughs will come in entertainment and social interaction. That’s where the endgame lies.
36Kr: Is there an AI services ecosystem emerging for glasses makers?
YL: Yes, but it’s still early. Building an agent is much easier than a traditional app. It’s more like a mini program. Most of the work is being done by startups and indie developers. We’ve committed RMB 50 million to build out this developer ecosystem and expect to see some fun, useful AI agents emerge.
36Kr: Which large models are you integrating into your product?
YL: We’re taking a hybrid approach. We use different models for different tasks: Doubao for chat, DeepSeek for search, and China Mobile’s Jiutian model for interactive scenarios.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This article was written by Qiu Xiaofen for 36Kr.